TITLE:
Calibration and Validation of Decision Support System for Agro-Technology Transfer Model for Simulating Growth and Yield of Maize in Bangladesh
AUTHORS:
F. Ahmed, Apurba K. Choudhury, S. Akhter, M. A. Aziz, Jatish C. Biswas, M. Maniruzzaman, M. Main Uddin Miah, M. M. Rahman, M. A. H. S. Jahan, Imrul Mosaddek Ahmed, R. Sen, S. Ishtiaque, A. F. M. Tariqul Islam, M. M. Haque, M. Belal Hossain, Naveen Kalra, M. Hafijur Rahman
KEYWORDS:
Calibration, Validation, DSSAT Model and Maize
JOURNAL NAME:
American Journal of Plant Sciences,
Vol.8 No.7,
June
23,
2017
ABSTRACT: Maize
is an emerging important crop in Bangladesh because of its high yield potential
and economic profitability compared to rice and wheat crops. There is a need to
understand the growth and yield behavior of this crop in varying production
environments of Bangladesh. Crop model such as Decision Support System For
Agro-technology Transfer (DSSAT) version 4.6 (DSSAT hereafter) can be utilized
cost effectively to study the performances of maize under different production
environments. It needs to calibrate and validate DSSAT model for commonly
cultivated maize cultivars in Bangladesh and subsequently take the
model to various applications, including inputs and agronomic management
options and climate change that impacts
analyses. So, the present study was undertaken to firstly calibrate DSSAT model
for popular four hybrid maize cultivars (BARI Hybrid Maize-7, BARI Hybrid
Maize-9, Pioneer 30B07 and NK-40).
Subsequently, it proceeded with the validation with independent field data sets
for evaluating their growth performances. The genetic coefficients for these
cultivars were evaluated by using Genotype coefficient calculator (GENCALC) and
Generalized likelihood uncertainty estimation (GLUE) module of DSSAT on the
basis of first season experiment. The performance of the model was satisfactory
and within the significant limits. After calibration, the model was tested for
its performance through validation procedure by using second season data. The
model performed satisfactorily through phenology, biomass, leaf area index
(LAI) and grain yield. Phenology, as estimated through days to flower
initiation and maturity, was in
good agreement, although simulated results were slightly over predicted compared
to observed values but within the statistical significance limit...when
compared with observed values at specific growth stages of the crop. The final
yield values (10.12 to 10.59 t·ha-1) were
in close agreement with the observed values (10.16 to 10.94 t·ha-1), as
the percentage error was within tolerable limit (0.39% to 6.81%). The model has
been successfully calibrated and validated for Gazipur environment and now can
be used for climate change impact studies for similar environments in
Bangladesh.